PDR Energy Baseline Alternative Proposal for Discussion October 27, 2015
Update Since Last Meeting Since the October 12 th Working Group Meeting Stem, Solar City, AMS and CESA have had over a half-dozen meetings to discuss adjustments to the CAISO’s proposed methods for estimating generation output used for retial purposes. We concluded that the CAISO’s proposed option titled “using a baseline” that calculates a Gt adjusted value is the most viable option if we must move forward quickly. Similar to the CAISO’s Gt adjusted construct, we propose establishing a “typical - usage level” (G TYPICAL ) based on electric output using (G) metered quantities during non-dispatch days. We recommend moving forward with a Gt adjusted calculation, while also continuing to discuss broader issues around dual- use in Phase II of the CAISO’s ESDER Initiative.
Background | Why do we need a change? In its recent Local Capacity Resource Procurement, SCE signed PPAs for 135 of Demand Response Energy Storage (DRES) Resources - 85 MW with Stem, 50 MW with AMS. Projects are being installed now, with testing occurring in the second half of 2015 and commercial operation dates of 10/1/2016. AMS contracts require use of an energy baseline unless there is a CAISO approved alternative. Resources are designed to be dispatched often, making use of the PDR baseline / collecting 10 days of accurate load data on non-event days could be challenging during certain times of the year. 3
Alternative Baseline| Use Case Location: Behind-the-Meter C&I installation Product: Load drop to an IOU ~100 times a year using DR and/or battery storage Value Streams: – Wholesale: Contractual dispatches by IOU – Retail: Energy arbitrage, demand charge management, retail demand response programs 4
Alternative Baseline| Proposal Overview Use the battery meter to directly measure/settle contractual dispatches; with an adjustment calculating out typical retail use. Similar to the CAISO’s Gt adjusted construct, establish a “typical - usage level” (G TYPICAL ) based on electric output using (G) metered quantities during non-dispatch days. Days would be selected by conducting a look back of comparable, non-dispatch days (similar to the CAISO’s existing 10 -in-10 in 45 baseline methodology for loads). Scenario A: Similar to the PDR baseline methodology, the first 10 non-event similar day-type days are selected. A non-event day is a day in which no hour contains a PDR dispatch. A profile is created by averaging 5-minute data to the hour and then over the 10 days. Performance (P) during each interval (t) during the PDR call is therefore: P = - (G - G TYPICAL ). Scenario B: If 10 non-event days cannot be found within a look-back window, then the resource is deemed to be a full-time market participant and G TYPICAL = 0 and the resource receives credit for the full generation value during the event duration (standard proposed PDR rules for MGO). 5
MGO Hour-Specific Adjustment Method Average of 10 non-event days over past 45 days Sept 22, 2014 Dispatch, 10 Prior Workdays for Adjustment Calculation Performance with 10-day Adjustment on MGO 1000 Energy (kWh) 750 500 Dispatch 250 0 -250 0:00 2:00 4:00 6:00 8:00 10:0012:0014:0016:0018:0020:0022:00 Facility load (kW) Site meter (kW) Performance, G_adj (kW) MGO adjustment 18% adjustment 6
Measurements at HE 15:00 in kW Meter 9/8 9/9 9/10 9/11 9/12 N 723 kW 723 kW 723 kW 719 kW 723 kW Meter Dispatch Day: 9/22 G -28 kW -28 kW -15 kW -8 kW -27 kW N 543 kW L 751 kW 751 kW 738 kW 727 kW 751 kW G -211 kW G_typical -37 kW Meter 9/15 9/16 9/17 9/18 9/19 L 754 kW N 723 kW 723 kW 723 kW 723 kW 723 kW P = -(G - G_typical) 174 kW G -127 kW -66 kW -50 kW -21 kW 0 kW L 851 kW 789 kW 773 kW 743 kW 700 kW 7
Alternative Baseline | Likely Dispatch Net Benefits Test Prices ($/MWh) and Workdays with LMP Above On Peak Price Oct-14 Nov-14 Dec-14 Jan-15 Feb-15 Mar-15 Apr-15 May-15 Jun-15 Jul-15 Aug-15 Sep-15 on peak 36.49 34.93 79.92 56.98 51.73 34.31 34.52 33.48 36.95 32.02 35.29 38.3 off peak 37.22 35.52 71.53 57.54 54.06 35.4 34.79 33.97 37.59 32.4 36.04 39.15 Dispatch 23 20 0 0 0 21 22 11 Days Oct-13 Nov-13 Dec-13 Jan-14 Feb-14 Mar-14 Apr-14 May-14 Jun-14 Jul-14 Aug-14 Sep-14 on peak 52.4 50.15 42.8 56.63 65.57 79.43 71.51 65.35 56.08 55.57 62.97 58.45 off peak 56.16 51.62 43.54 57.42 66.23 80.26 72.79 67.86 61.95 64.09 64.38 64.21 Dispatch 1 4 22 8 11 0 0 1 0 0 0 0 Days Oct-12 Nov-12 Dec-12 Jan-13 Feb-13 Mar-13 Apr-13 May-13 Jun-13 Jul-13 Aug-13 Sep-13 on peak 31.87 39.44 40.53 37.25 37.18 38.05 41.63 45.93 48.6 47.79 44.64 41.35 off peak 32.13 39.38 41.14 38.06 37.98 39.23 42.66 48.21 49.9 51.46 48.22 42.19 Dispatch 23 15 12 23 20 21 22 13 6 8 9 16 Days 8
Additional Slides (If Needed)
Most Recent 10 Days without Dispatch February 17-28, 2014: one week has 4 days of peak shaving 10
Two Previous Weeks: No Peak Shaving 11
Recommend
More recommend
